The goal of the proposed research is a detailed analysis of the export machinery for P. aeruginosa exotosin A, including further tests of our model in which proteins are excreted by lateral movement through the Bayer junction, rather them by secretion across both membranes. Specific features of that model will be tested by developing an in vitro translocation system, in which these proteins are synthesized in the presence of inverted membrane vesicles. We will examine the kinetics of disulfide bond formation during excretion and will identify and localize enzymes involved in this process. We will also generate monoclonal antibodies against exotoxin A and pilin, in order to identify epitopes involved in various steps of export to map the domains of exotoxin A involved in catalysis of the ADP-ribosyl transfer and in cytotoxicity, and to study posttranslational modifications of pilin as well as interactions of pili with eukaryotic cells. Using monoclonal antibodies against the signal sequence fo exotoxin A and pilin, anti-idiotype antibodies will be prepared. These should resemble in their specificity the signal, and they may be useful in identifying and isolating the bacterial components that recognize the signal, along with associated additional components of the export machinery. Mutants defective in export of exotoxin A and pilin will be isolated, and these will be used by complementation to further identify components of an apparatus of secretion. Finally, attempts will be made to clone, in E. coli, the structural genes for pilin and exotoxin A, and genes involved in their export; and the recombinant plasmids will be used to sequence the signal, to identify possible addtional hydrophobic domains, and alter export, by site specific mutagenesis.